Stage-aware scoring: why a $0 MVP shouldn't get the same audit as a Series A
Stage-aware scoring is the practice of redistributing audit-lens weights based on the company's stage (idea, mvp, seed, series_a_plus) — early stages weight Problem-Solution and Team heavily; later stages weight Traction and Financials heavily.
A stage-aware audit redistributes the 6 lens weights by company stage. At idea-stage Problem-Solution gets 35% of the score; at Series A+ that drops to 15% and Traction takes 30%. Same idea, same data, different stage = different verdict. This catches the failure mode where investors say 'great idea but too early' or 'great traction but too late' — the score now reflects which one applies.
Key facts
- Idea-stage weights: Problem-Solution 35%, Team 25%, Market 15%, Competition 15%, GTM 5%, Financials 5%.
- Seed weights: Team 25%, Traction 25%, Problem-Solution 20%, Competition 15%, GTM 10%, Financials 5%.
- Series A+ weights: Traction 30%, Financials 25%, Team 15%, GTM 15%, Competition 10%, Problem-Solution 5%.
- Same idea audited at idea-stage vs Series A produces verdict deltas of 25-40 points in 60% of test runs.
Most "idea validators" run the same scoring rubric whether you're an idea on a napkin or a $5M ARR Series A. That's why their verdicts feel disconnected from reality — a Series A operator gets the same "47/100, Yellow zone" as a hobbyist with a one-line idea.
Methodology v2 fixes this by redistributing the 6-lens weights based on stage.
The four stages
| Stage | Definition |
|---|---|
| idea | No product, no users, just a thesis. |
| mvp | Built product, < $1k MRR or < 100 paying users. |
| seed | $1k-100k MRR, looking to raise pre-A. |
| series_a_plus | $100k+ MRR, scaling team and motion. |
The weight redistribution
| Lens | idea | mvp | seed | series_a+ |
|---|---|---|---|---|
| Team | 25% | 25% | 25% | 15% |
| Problem-Solution | 35% | 30% | 20% | 5% |
| Traction | 0% | 10% | 25% | 30% |
| Competition | 15% | 15% | 15% | 10% |
| GTM | 5% | 5% | 10% | 15% |
| Financials | 5% | 5% | 5% | 25% |
| Market | 15% | 10% | 0% | 0% |
(Yes, weights add to 100% per stage. Market is folded into Problem-Solution at later stages — at Series A nobody re-debates the market, you've proven it.)
Why these numbers
Three principles:
- Risk shifts as the company matures. Idea-stage risk is "is the problem real?" — so Problem-Solution carries the score. Series A risk is "can you scale?" — so Traction and Financials carry it.
- What you can't fake at this stage gets the most weight. Idea-stage founders can't fake good problem research — so we test it. Series A founders can't fake net-new-revenue numbers — so we test those.
- Cross-stage gaming doesn't pay. Inflating your stage to escape harsh Problem-Solution scrutiny just moves you into Traction-weighted scoring you can't fake either.
What this looks like in practice
Same idea — "AI note-taking app for lawyers" — with different stage signals:
- Idea-stage: Problem-Solution 78/100 (good Reddit signal), Team 65 (founders are software engineers, not lawyers), Market 70. Composite: 65 / Yellow / REFINE.
- Series A+: Traction 88 (real paying law firms, $250k MRR, 18% MoM), Financials 75, Team 80 (now you have the head of legal-tech eng from Clio). Composite: 84 / Green / GO.
Same idea. Different verdicts. The first audit is honest about idea-stage uncertainty. The second is honest about traction-stage strength. Neither would be useful if both used the same weighting.
How to run it
audit this idea: <your idea> --stage idea
audit this idea: <your idea> --stage seed
Or let extract_idea auto-classify based on the signals in your prompt. The extract step is right ~85% of the time and you can override the rest.
curl -fsSL https://inite.studio/install.sh | sh
FAQ
Frequently asked questions
Why does Problem-Solution drop to 5% at Series A+?
By Series A you've proven the problem exists — your traction shows that. The remaining risk shifts to scaling, unit economics, and team execution. Re-debating Problem-Solution at Series A is what bad audits do.What if I don't know what stage I'm at?
The extract_idea step classifies stage automatically based on signals: revenue + paying customers = seed/A; built product no revenue = mvp; just an idea = idea. You can override but the auto-classification is right ~85% of the time in our dataset.Does this mean idea-stage audits are easier to GO?
No — the bar moves with the weights. Idea-stage demands rigorous problem evidence (real Reddit complaints, SERP saturation analysis). 'Plausible idea' isn't enough. The system errs toward false negatives at every stage.Can I run the same audit at multiple stages to see how my score evolves?
Yes. Re-run with `--stage seed` then `--stage series_a_plus`. The dashboard shows the delta. Useful for planning: what's the score I'll have when I'm raising next round, given current trajectory?